Solvation Enthalpy Determination for Aqueous-Phase Reaction Adsorbates Using Ab Initio Molecular Dynamics-Based Structure Sampling

Abstract

The interplay between covalent and noncovalent interactions at the solid−liquid interface strongly influences electrocatalytic reactions. Although methods to determine the former interactions have been rigorously developed, the latter are often described with static bilayer models or similarly approximate methods. In this study, we account for the disorder and dynamics at complex electrochemical interfaces by proposing a simple theory to estimate the enthalpy of solvation for adsorbed intermediates. In a strategy reminiscent of Born−Haber cycles, the enthalpy of solvation is expressed in terms of two subprocesses: vacancy creation by water reorganization and adsorbate interaction with solvent molecules. The magnitude of the solvation enthalpy is then determined as a mean valueof hydrated adsorbate-catalyst configurations obtained from simulated annealing with ab initio molecular dynamic (AIMD) simulations. This theory can be generalized for many combinations of surfaces and adsorbates. Its application improves treatment of energetics at electrified solid−liquid interfaces as well as corresponding structure−activity−stability predictions, as demonstrated for electrochemical oxygen reduction on Pt(111) and on Fe−N−C catalysts and for ethanol electrooxidation on Pt(111).

Toward Realistic Models of the Electrocatalytic Oxygen Evolution Reaction

“…While the effects of the electric field have been shown, at least in some cases, to be small, solvation effects cannot be ignored, since different reaction intermediates interact differently with water-based electrolytes. These interactions can be as large as 0.5–0.6 eV for intermediates like *OOH and *OH that form multiple hydrogen bonds with water, while they can be as low as 0.1 eV for *O, which can interact weakly with water. , Implicit PCM-type solvent models are unable to describe directional interactions like H-bonds and are therefore of limited value in this regard . An explicit description of water, with solvation effects obtained from statistical averages of structures extracted from molecular dynamic simulations, offers a viable, albeit time-consuming, way to estimate solvation effects. , …”

“…Similarly to the work of Wang et al, the strategy adopted by Chun et al in their very recent work is also based on the structural optimization of several snapshots extracted from FPMD simulations of various reaction intermediates with an explicit description of the water solvent. Focusing on the ORR on Pt(111), the authors estimated the solvation energy for the *OH and *OOH intermediates to be −0.56 eV and −0.27 eV, respectively.…”